Abstract: An architecture to allow Multi-Access Edge Computing (MEC) billing and charge tracking, is disclosed. In an example, a tracking process, such as is performed by an edge computing apparatus, includes: receiving a computational processing request for a service operated with computing resources of the edge computing apparatus from a connected edge device within the first access network, wherein the computational processing request includes an identification of the connected edge device; identifying a processing device, within the first access network, for performing the computational processing request; and storing the identification of the connected edge device, a processing device identification, and data describing the computational processes completed by the processing device in association with the computational processing request.

Abstract: Various approaches for deployment and use of configurable edge computing platforms are described. In an edge computing system, an edge computing device includes hardware resources that can be composed from a configuration of chiplets, as the chiplets are disaggregated for selective use and deployment (for compute, acceleration, memory, storage, or other resources). In an example, configuration operations are performed to: identify a condition for use of the hardware resource, based on an edge computing workload received at the edge computing device; obtain, determine, or identify properties of a configuration for the hardware resource that are available to be implemented with the chiplets, with the configuration enabling the hardware resource to satisfy the condition for use of the hardware resource; and compose the chiplets into the configuration, according to the properties of the configuration, to enable the use of the hardware resource for the edge computing workload.

Abstract: Various approaches for implementing platform resource management are described. In an edge computing system deployment, an edge computing device includes processing circuitry coupled to a memory. The processing circuitry is configured to obtain, from an orchestration provider, an SLO (or SLA) that defines usage of an accessible feature of the edge computing device by a container executing on a virtual machine within the edge computing system. A computation model is retrieved based on at least one key performance indicator (KPI) specified in the SLO. The defined usage of the accessible feature is mapped to a plurality of feature controls using the retrieved computation model. The plurality of feature controls is associated with platform resources of the edge computing device that are pre-allocated to the container. The usage of the platform resources allocated to the container is monitored using the plurality of feature controls.

Abstract: Systems and techniques for AI model and data camouflaging techniques for cloud edge are described herein. In an example, a neural network transformation system is adapted to receive, from a client, camouflaged input data, the camouflaged input data resulting from application of a first encoding transformation to raw input data. The neural network transformation system may be further adapted to use the camouflaged input data as input to a neural network model, the neural network model created using a training data set created by applying the first encoding transformation on training data. The neural network transformation system may be further adapted to receive a result from the neural network model and transmit output data to the client, the output data based on the result.

Abstract: Software and other electronic services are increasingly being executed in cloud computing environments. Edge computing environments may be used to bridge the gap between cloud computing environments and end-user software and electronic devices, and may implement Functions-as-a-Service (FaaS). FaaS may be used to create flavors of particular services, a chain of related functions that implements all or a portion of a FaaS edge workflow or workload. A FaaS Temporal Software-Defined Wide-Area Network (SD-WAN) may be used to receive a computing request and decompose the computing request into several FaaS flavors, enable dynamic creation of SD-WANs for each FaaS flavor, execute the FaaS flavors in their respective SD-WAN, return a result, and destroy the SD-WANs. The FaaS Temporal SD-WAN expands upon current edge systems by allowing low-latency creation of SD-WAN virtual networks bound to a set of function instances that are created to a execute a particular service request.

Abstract: A system for trust brokering as a service includes an edge computing node and a trust brokering service edge computing device. The trust brokering service edge computing device receives a computing workload request from an application configured to process secure data and identifies a set of security requirements associated with the request. The device also identifies a security feature present in the set of security requirements but not provided by the edge computing node. To address this, the device generates an application execution environment that includes a secure plugin providing the security feature and a virtual device representing the edge computing node. The computing workload request is then executed at the application execution environment, providing a secure and efficient solution for trust brokering as a service.

Abstract: An apparatus to facilitate provenance audit trails for microservices architectures is disclosed. The apparatus includes one or more processors to: obtain, by a microservice of a service hosted in a datacenter, provisioned credentials for the microservice based on an attestation protocol; generate, for a task performed by the microservice, provenance metadata for the task, the provenance metadata including identification of the microservice, operating state of at least one of a hardware resource or a software resource used to execute the microservice and the task, and operating state of a sidecar of the microservice during the task; encrypt the provenance metadata with the provisioned credentials for the microservice; and record the encrypted provenance metadata in a local blockchain of provenance metadata maintained for the hardware resource executing the task and the microservice.

Abstract: Various systems and methods for providing consensus-based named function execution are described herein. A system is configured to access an interest packet received from a user device, the interest packet including a function name of a function and a data payload; broadcast the interest packet to a plurality of compute nodes, wherein the plurality of compute nodes are configured to execute a respective instance of the function; receive a plurality of responses from the plurality of compute nodes, the plurality of responses including respective results of the execution of the respective instances of the function; analyze the plurality of responses using a consensus protocol to identify a consensus result; and transmit the consensus result to the user device.

Abstract: Telemetry systems for monitoring cooling of compute components and related apparatus and methods are disclosed. An example apparatus includes interface circuitry, machine-readable instructions, and programmable circuitry to at least one of instantiate or execute the machine-readable instructions to generate a heatmap based on outputs of one or more sensors in an environment, the environment including a first compute device, the sensor outputs including a metric associated with a property of a coolant and a location of the sensor in the environment, identify a compute performance metric of the first compute device, determine a cooling parameter for the first compute device based on the heatmap and the compute performance metric, and cause a cooling distribution unit to control flow of the coolant in the environment based on the cooling parameter.

Abstract: Systems, apparatus, and methods for managing cooling of compute components are disclosed. An example apparatus includes programmable circuitry to at least one of instantiate or execute machine readable instructions to identify a workload to be performed by a compute device, identify a service level objective associated with the workload or the compute device, determine a parameter of a coolant to enable the service level objective to be satisfied during performance of the workload, and cause a cooling distribution unit to control the coolant based on the coolant parameter.

Abstract: Various systems and methods for implementing a service-level agreement (SLA) apparatus receive a request from a requester via a network interface of the gateway, the request comprising an inference model identifier that identifies a handler of the request, and a response time indicator. The response time indicator relates to a time within which the request is to be handled indicates an undefined time within which the request is to be handled. The apparatus determines a network location of a handler that is a platform or an inference model to handle the request consistent with the response time indicator, and routes the request to the handler at the network location.

Abstract: Various approaches for implementing multi-tenant data protection are described. In an edge computing system deployment, a system includes memory and processing circuitry coupled to the memory. The processing circuitry is configured to obtain a workflow execution plan that includes workload metadata defining a plurality of workloads associated with a plurality of edge service instances executing respectively on one or more edge computing devices. The workload metadata is translated to obtain workload configuration information for the plurality of workloads. The workload configuration information identifies a plurality of memory access configurations and service authorizations identifying at least one edge service instance authorized to access one or more of the memory access configurations. The memory is partitioned into a plurality of shared memory regions using the memory access configurations.

Abstract: A service coordinating entity device includes communications circuitry to communicate with a first access network, processing circuitry, and a memory device. The processing circuitry is to perform operations to, in response to a request for establishing a connection with a user equipment (UE) in a second access network, retrieve a first Trusted Level Agreement (TLA) including trust attributes associated with the first access network. One or more exchanges of the trust attributes of the first TLA and trust attributes of a second TLA associated with the second access network are performed using a computing service executing on the service coordinating entity. A common TLA with trust attributes associated with communications between the first and second access networks is generated based on the exchanges. Data traffic is routed from the first access network to the UE in the second access network based on the trust attributes of the common TLA.

Abstract: Various systems and methods of initiating and performing contextualized AI inferencing, are described herein. In an example, operations performed with a gateway computing device to invoke an inferencing model include receiving and processing a request for an inferencing operation, selecting an implementation of the inferencing model on a remote service based on a model specification and contextual data from the edge device, and executing the selected implementation of the inferencing model, such that results from the inferencing model are provided back to the edge device. Also in an example, operations performed with an edge computing device to request an inferencing model include collecting contextual data, generating an inferencing request, transmitting the inference request to a gateway device, and receiving and processing the results of execution. Further techniques for implementing a registration of the inference model, and invoking particular variants of an inference model, are also described.

Abstract: Logic may determine a specific performance of a neural network based on an event and may present the specific performance to provide a user with an explanation of the inference by a machine learning model such as a neural network. Logic may determine a first activation profile associated with the event, the first activation profile based on activation of nodes in one or more layers of the neural network during inference to generate an output. Logic may correlate the first activation profile against a second activation profile associated with a first training sample of training data. Logic may determine that the first training sample is associated with the event based on the correlation. Logic may output an indicator to identify the first training sample as being associated with the event.

Abstract: Embodiments described herein are generally directed to intelligent management of microservices failover. In an example, responsive to an uncorrectable hardware error associated with a processing resource of a platform on which a task of a service is being performed by a primary microservice, a failover trigger is received by a failover service. A secondary microservice is identified by the failover service that is operating in lockstep mode with the primary microservice. The secondary microservice is caused by the failover service to takeover performance of the task in non-lockstep mode based on failover metadata persisted by the primary microservice. The primary microservice is caused by the failover service to be taken offline.

Abstract: Embodiments described herein are generally directed to the use of sidecars to perform dynamic API contract generation and conversion. In an example, a first call by a first microservice to a first API of a second microservice is intercepted by a first sidecar of the first microservice. The first API is of a first API type of multiple API types and is specified by a first contract. An API type of the multiple API types is selected by the first sidecar. Responsive to determining the selected API type differs from the first API type, based on the first contract, a second contract is generated by the first sidecar specifying a second API of the selected API type; and a second sidecar of the second microservice is caused to generate the second API and internally connect the second API to the first API based on the second contract.

Abstract: An apparatus to facilitate provenance audit trails for microservices architectures is disclosed. The apparatus includes one or more processors to: obtain, by a microservice of a service hosted in a datacenter, provisioned credentials for the microservice based on an attestation protocol; generate, for a task performed by the microservice, provenance metadata for the task, the provenance metadata including identification of the microservice, operating state of at least one of a hardware resource or a software resource used to execute the microservice and the task, and operating state of a sidecar of the microservice during the task; encrypt the provenance metadata with the provisioned credentials for the microservice; and record the encrypted provenance metadata in a local blockchain of provenance metadata maintained for the hardware resource executing the task and the microservice.

Abstract: An apparatus to facilitate matchmaking-based enhanced debugging for microservices architectures is disclosed. The apparatus includes one or more processors to: detect, by an anomaly detector in a sidecar of a microservice hosted by a container, an anomaly in telemetry data generated by the microservice, the microservice hosted in a container executed by the processor and part of a service of an application; enable, by an enhanced debug and trace component of the sidecar, a debug mode in the microservice, the debug mode based on a type of the anomaly; collect, by the enhanced debug and trace component, a target set of data points generated by the microservice; and process, by the enhanced debug and trace component, the target set of data points with a matchmaking process to generate a timestamp and a tag for a context for each data point of the target set of data points.

Abstract: An apparatus to facilitate at-scale telemetry using interactive matrix for deterministic microservices performance is disclosed. The apparatus includes one or more processors to: receive user input comprising an objective or task corresponding to scheduling a microservice for a service, wherein the objective or task may include QoS, SLO, ML feedback; identify interaction matrix components in an interaction matrix that match the objective or tasks for the microservice; identify knowledgebase components in knowledgebase that match the objective or tasks for the microservice; and determine a scheduling operation for the microservice, the scheduling operation to deploy the microservice in a configuration that is in accordance with the objective or task, wherein the configuration comprises a set of hardware devices and microservice interaction points determined based on the interaction matrix components and the knowledgebase components.